Determination of arsenic species in fish oil after acid digestion

Arsenic speciation is of increasing interest to the food industry, as concerns about high total arsenic concentrations in food can often be alleviated to a great extent if the ratio of toxic, less toxic and non-toxic arsenic compounds in the sample is known. The lipid matrix of fish oil is a challenge in the determination of arsenic species, as current methods for this type of analysis require the analyte to be water-soluble. In this study, two sample preparation techniques were applied. One the one hand water-soluble species were extracted with methanol/water, on the other, acid digestion was applied to release lipid-soluble arsenic compounds into the aqueous phase. Ion chromatography - inductively coupled plasma mass spectrometry (IC-ICP-MS) was used for separation and sensitive element-specific detection of arsenic compounds. Additional experiments, including alkaline hydrolysis, were carried out to find out more about the type of lipids arsenic is bound to in fish oil. Up to eight different arsenic species were detected and quantified in fish oil with dimethylarsinate being the major compound both in the aqueous extract and in the acid digest. No inorganic arsenic was detected in the aqueous extract, and the maximum concentration of arsenate determined in the acid digest was 0.05 mu g g(-1). The total arsenic concentration determined by ICP-MS ranged from < 0.1 to 5 mu g g(-1) supercript stop. With regard to the mass balance, approximately 1% of the total arsenic content was extractable with methanol/water, whereas the sum of arsenic species quantified after acid digestion yielded 85-100% of the total arsenic content. It was confirmed that the large fraction of arsenic in fish oil not extractable on an aqueous basis consists of organoarsenic compounds. This new approach in sample preparation makes the complete characterization of the arsenic content in the sample possible with regard to the respective species, providing necessary information required for risk assessment.